SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Linux isn't UNIX, it's UNIX-like, and GNU isn't UNIX, it's a software package collection of tools that can function as a core operating system with any type of kernel behind it... including Linux, BSD, UNIX, or even DOS.
GNU was meant to be an open drop-in replacement for the standard UNIX toolkits and operating system environments, and Linux was a drop-in replacement for UNIX and UNIX-like kernels. In fact GNU actually has it's own kernel, HURD, but because HURD was never completed, most GNU operating systems ended up as GNU/Linux operating systems. And also because 386/BSD was being sued and wasn't freely available, Linus had his hand forced to create the Linux kernel.
Every UNIX and UNIX-like kernel and operating system all share a commonality.
However a true GNU operating environment is supposed to be kernel-independent with only the shell and shell-core utilities recompiled between each kernel to interface and function with it. This way you can have GNU/Linux, GNU/HURD, GNU/BSD, etc. in any configuration if deemed applicable.
Didn't want to double post but my reply was a bit long, so this is only to break it down somewhat.
I could see where systemd MIGHT be useful for a power user or hardcore Linux enthusiast with a fully customized system who wants a more aggressive break-your-neck speed system, but servers, average, and new users need more simplified tools that don't require a lot of work and debugging.
Sysvinit and bsdinit are simple tools for simple people, and yes, systemd would break the KISS principle of Slackware completely. It's a fast system to use, but as far as reliability goes, it's sketchy, and it requires too much work to get working correctly, and it's system heavy is resource usage.
RHEL makes billions from servers. If switching to systemd will make RHEL's life harder it has much more to lose then any of us in here. I highly doubt they will adopt something that will cause such a problem. I trust RHEL's choices as much as I trust Patricks choices. If RHEL 7 does not adopt systemd, then I will also agree that systemd is a failure. But until then I am undecided.
##Off topic. I tried Kubuntu but realized it was Ubuntu + More Bugs so I switched back to Unity Shell which I think Ubuntu has done a much better job in terms of stability. I think SLackware pulls it off best with the KDE platform compared to the other distros I tried. Even though 13.37 uses an older version kde.
Last edited by Mercury305; 08-24-2012 at 01:51 AM.
Linux isn't UNIX, it's UNIX-like, and GNU isn't UNIX, it's a software package collection of tools that can function as a core operating system with any type of kernel behind it... including Linux, BSD, UNIX, or even DOS.
GNU was meant to be an open drop-in replacement for the standard UNIX toolkits and operating system environments, and Linux was a drop-in replacement for UNIX and UNIX-like kernels. In fact GNU actually has it's own kernel, HURD, but because HURD was never completed, most GNU operating systems ended up as GNU/Linux operating systems. And also because 386/BSD was being sued and wasn't freely available, Linus had his hand forced to create the Linux kernel.
Every UNIX and UNIX-like kernel and operating system all share a commonality.
However a true GNU operating environment is supposed to be kernel-independent with only the shell and shell-core utilities recompiled between each kernel to interface and function with it. This way you can have GNU/Linux, GNU/HURD, GNU/BSD, etc. in any configuration if deemed applicable.
GNU is a Communist Licence. Its strange and obtuse. However this strangeness gives corporations a comparative advantage using it. As it takes advantage of free labor of developers without having to pay them for it. It has different versions. For example the GNU Kernel is Version 2 while there is a version 3 being currently used by much of the tools of the GNU userland.
BSD is much more direct and practical. Its open source but it gives certain rights to the software developers. But since large corporations can't take advantage of the loopholes GNU has they rarely finance BSD as they do with GNU. Now you know why LInux grew so much and BSD died out.
I don't like extremes. It always is corrupt in 1 way or another. I agree with you systemd may target elimination of BSD/*NIX'es but BSD has pretty much self eliminated itself due to lack of corporate support. Don't get me wrong. What they do is great. A great practical working system. But without the support what use is it?
Thats just life. Its corrupt. We can either bark and shout endlessly getting no where or just move on. At least Slackware uses the Linux Kernel and has the ability to adapt to changes. But the same can not be said of BSD. Even if it isnt systemd... it would have ended being something else.
Business is business. Its about profit.
So yes, Linux can be quite hostile to other OS's with this comparative advantage of GNU License. But the beauty is that Linux distros are much less hurting each other by these changes. After all, they share a mutual Kernel and even the userland. It may not end up friendly with "Unix Philosophy" but at least its less self destructive towards other LInux distros.
Last edited by Mercury305; 08-24-2012 at 02:14 AM.
Systemd is going top need at least 2 years out in the mainstream before I would dare, in my opinion, consider it in usage and to allow it to prove itself and proof itself... and yes you read right.
The system resource usage and bloat needs to be trimmed also. Sysvinit and bsdinit's memory footprint is tiny compared to systemd's and on system with not a lot of resources it could be problematic, and even servers where tight and controlled resource handling is make or break, plus systemd needs to shed it's Linux only status and be more open to other kernels and systems.
To quote Pro Wrestler Ric Flair, "To be 'the man', you have to beat 'the man'." and that's in all fields, all formats, and each and every way. This means it has to be portable to BSD, and other UNIX systems to show it can replace sysvinit and bsdinit, and not just on Linux.
And Mercury, BSD isn't a system designed around a Home/Office/Laptop computer. It's a system designed for servers. The port PC-BSD does try to bring FreeBSD pre-configured to a desktop environment, but BSD is not an operating system targeted at general usage. The only reason it's behind in some areas of development like driver support is because the developers want to release stable code that works, not just a half-working or barely functional staging driver requiring tons of firmware.
If systemd had came out as a multi-UNIX platform init replacement for all kernels and all systems, it might have been looked on with greater respect.
The right choice of action for Slackware right now is clearly is the one Pat is taking, "Wait it out and see", given that Slackware prides itself on not taking unnecessary and risky changes. There is no major reason for Slackware to shift to systemd right now. The current init system works nicely and plays well with the other components as they stand today. systemd might be gaining ground fast but it is still unproven on critical systems that run distros like RHEL, Debian or even Slackware.
In the future because many distros are taking up systemd (and that likely includes RHEL7) two things will probably happen, systemd will have started to prove itself on some of these critical system (or problems it has will be ironed out) and more of other Linux OS components will begin to depend on systemd. At this stage thinking about systemd on Slackware might make some sense but right now it is very much a "why bother?". The only minor advantage Slackware might get (for the vast amount of work it would take to switch) is a possible speed up in boot times, with the disadvantage of higher risks of failures and problems due to unproven code.
Regarding systemd on Arch, initially I was surprised that Arch seemed to be rushing headlong into "systemd land" but actually the more you think about it, it does make more sense for them. Sure Arch and Slack do share many core beliefs (e.g. configuration via text files and vanilla packages) but some aspects of their philosophy are very different. Slackware focuses on stability and Arch focusses on allowing its users to test the new stuff in the Linux world. Like it or loath it systemd fits that description (it is one of the big, new things that people are talking about) and because Arch cycle though new packages so fast, they are more likely to hit problems with other stuff depending on it sooner than Slackware. So sure, this might be the right time for Arch but it isn't for Slackware.
In summary I can't help but think that this thread, whilst interesting is at least a couple (if not more) years too early. systemd might just come to Slackware one day (particularly if almost everything else eventually depends on it) but now it is too early to spend any real time worrying about it.
P.S. There remains the possibility that systemd will fall flat on its face and then Pat's "wait and see" approach will start to look very wise, even to those outside the Slackware community (who were already acutely aware of it). And even if it turns out to be an amazing success, I still contend that waiting does no harm.
Last edited by ruario; 08-24-2012 at 03:32 AM.
Reason: changed one instance of 'might' to 'would'
I switch my computer off at night (or in the early hours of the morning) before going to bed, and switch it on when I get up. It boots up, becomes fully functional, a hell of a lot faster than I do. It's me that needs systemd, not my Slackware.
The right choice of action for Slackware right now is clearly is the one Pat is taking, "Wait it out and see", given that Slackware prides itself on not taking unnecessary and risky changes. There is no major reason for Slackware to shift to systemd right now. The current init system works nicely and plays well with the other components as they stand today. systemd might be gaining ground fast but it is still unproven on critical systems that run distros like RHEL, Debian or even Slackware.
In the future because many distros are taking up systemd (and that likely includes RHEL7) two things will probably happen, systemd will have started to prove itself on some of these critical system (or problems it has will be ironed out) and more of other Linux OS components will begin to depend on systemd. At this stage thinking about systemd on Slackware might make some sense but right now it is very much a "why bother?". The only minor advantage Slackware might get (for the vast amount of work it would take to switch) is a possible speed up in boot times, with the disadvantage of higher risks of failures and problems due to unproven code.
Regarding systemd on Arch, initially I was surprised that Arch seemed to be rushing headlong into "systemd land" but actually the more you think about it, it does make more sense for them. Sure Arch and Slack do share many core beliefs (e.g. configuration via text files and vanilla packages) but some aspects of their philosophy are very different. Slackware focuses on stability and Arch focusses on allowing its users to test the new stuff in the Linux world. Like it or loath it systemd fits that description (it is one of the big, new things that people are talking about) and because Arch cycle though new packages so fast, they are more likely to hit problems with other stuff depending on it sooner than Slackware. So sure, this might be the right time for Arch but it isn't for Slackware.
In summary I can't help but think that this thread, whilst interesting is at least a couple (if not more) years too early. systemd might just come to Slackware one day (particularly if almost everything else eventually depends on it) but now it is too early to spend any real time worrying about it.
P.S. There remains the possibility that systemd will fall flat on its face and then Pat's "wait and see" approach will start to look very wise, even to those outside the Slackware community (who were already acutely aware of it). And even if it turns out to be an amazing success, I still contend that waiting does no harm.
I am in 100% agreement with you.
Besides systemd.
I personally am surprised to come across more problems with Ubuntu then I do with Fedora even though Fedora is a testing distro. I guess this proves the superiority of its developers over Ubuntu. I have to give Fedora Developers some credit. It is pretty hard to get a distro to work with all the jibberish and testing they add into it. And if they can pull it off better then Ubuntu then that should prove they are better. I have never been a fan of Debian systems since 1998.
Slackware was number 1 and RH was number 2 for me back then and Still is.
I think a wise business decision would be for Slackware to use Arch the same way that RHEL uses Fedora for all the "new and cool stuff". Why? because its closer to Slackware Core.
I am mad that all the best software repos are directed towards this stupid Ubuntu distro. I did an update last night and my computer got killed. So I'm back on Slackware. I am actually thankful this happened.
Last edited by Mercury305; 08-24-2012 at 08:37 AM.
Systemd is going top need at least 2 years out in the mainstream before I would dare, in my opinion, consider it in usage and to allow it to prove itself and proof itself... and yes you read right.
The system resource usage and bloat needs to be trimmed also. Sysvinit and bsdinit's memory footprint is tiny compared to systemd's and on system with not a lot of resources it could be problematic, and even servers where tight and controlled resource handling is make or break, plus systemd needs to shed it's Linux only status and be more open to other kernels and systems.
To quote Pro Wrestler Ric Flair, "To be 'the man', you have to beat 'the man'." and that's in all fields, all formats, and each and every way. This means it has to be portable to BSD, and other UNIX systems to show it can replace sysvinit and bsdinit, and not just on Linux.
And Mercury, BSD isn't a system designed around a Home/Office/Laptop computer. It's a system designed for servers. The port PC-BSD does try to bring FreeBSD pre-configured to a desktop environment, but BSD is not an operating system targeted at general usage. The only reason it's behind in some areas of development like driver support is because the developers want to release stable code that works, not just a half-working or barely functional staging driver requiring tons of firmware.
If systemd had came out as a multi-UNIX platform init replacement for all kernels and all systems, it might have been looked on with greater respect.
If you are that keen on stability then may I ask why you are using Arch? A system that has rolling release seems crazy on a stability standpoint. The updates will wreck more havoc then systemd will I can guarantee you that from personal usage on even more tested distros like Ubuntu and its frequent updates.
As for BSD you are limiting the reasons why it is failing in both workstation and server market. Yes, i know its for a specific audience. But why do servers no longer adopt it as much as they used to? Not everything is a conspiracy here. If you do some research on it you will see the reason. Its not only GNU but they have a problem on the inside as well. A lot of top notch developers quit the project. I'm not saying it is going to die. But its kinda wounded pretty badly right now. ZFS seems to be their top reason for keeping some of their users. It may re emerge sometime in the future when Linux as a whole f***s up. I think Lennart was right when he said FreeBSD was no longer relevant anymore. As much as my thought will piss off the majority of BSD users in here. But on the bright side: It may not be relevant today but maybe in the future it may surpass Linux... Who knows? Time will tell. A lot of BSD users have switched to Slackware or even Arch. From what I here Arch even has a ports system.
FreeBSD was the Slackware of Linux back in the late 1990s. Linux is not UNIX you said it yourself. Therefore Linux has no responsibility of cross platform systems. The only way FreeBSD to succeed is to build its self its own way. If it continues to be dependent on Linux then I would not be too confident on FreeBSDs future.
If you are that keen on stability then may I ask why you are using Arch? A system that has rolling release seems crazy on a stability standpoint. The updates will wreck more havoc then systemd will I can guarantee you that from personal usage on even more tested distros like Ubuntu and its frequent updates.
Oh, so you can 'guarantee' this can you despite the fact that you admit you have never used it? WTF!? I used Arch as my primary work and home distro for a year and a half, perhaps more. Yes, it is rolling release but the development team do an amazing job of keeping it stable, whilst bringing out updates at a breakneck pace. It generally gave me less problems than other distros I had used up to that point. Yes, I had one or two minor hiccups after big upgrades but nothing that wasn't easy to resolve with just a little reading on the Arch forums or Wiki.
I wouldn't assume that just because a distro updates rapidly it is totally unstable. Your experience on Ubuntu or Fedora means nothing with regards to Arch. The only way to know how Arch (or any distro for that matter) really is, is to test it.
Just to be clear, I switched to Slackware not because of perceived stability improvements (though yes, after using it for a while now I would say Slackware is better in this regard) but rather because I didn't need the bleeding edge, 'latest and greatest' packages and I'd prefer not spend the bandwidth constantly downloading updates I didn't really care about, nor the time reviewing updates to config files (for the most part you can simply ignore them but it is nice to check them from time to time to see if there is anything you might want to consider merging in).
With regards to FreeBSD, once again I would tell you to actually try running it for a while before you go ahead and assume you have all the answers because frankly you seem to have very little of a clue about what the actual situation is.
Oh, so you can 'guarantee' this can you despite the fact that you admit you have never used it? WTF!? I used Arch as my primary work and home distro for a year and a half, perhaps more. Yes, it is rolling release but the development team do an amazing job of keeping it stable, whilst bringing out updates at a breakneck pace. It generally gave me less problems than other distros I had used up to that point. Yes, I had one or two minor hiccups after big upgrades but nothing that wasn't easy to resolve with just a little reading on the Arch forums or Wiki.
I wouldn't assume that just because a distro updates rapidly it is totally unstable. Your experience on Ubuntu or Fedora means nothing with regards to Arch. The only way to know how Arch (or any distro for that matter) really is, is to test it.
Just to be clear, I switched to Slackware not because of perceived stability improvements (though yes, after using it for a while now I would say Slackware is better in this regard) but rather because I didn't need the bleeding edge, 'latest and greatest' packages and I'd prefer not spend the bandwidth constantly downloading updates I didn't really care about, nor the time reviewing updates to config files (for the most part you can simply ignore them but it is nice to check them from time to time to see if there is anything you might want to consider merging in).
With regards to FreeBSD, once again I would tell you to actually try running it for a while before you go ahead and assume you have all the answers because frankly you seem to have very little of a clue about what the actual situation is.
You don't have to jump off a cliff to know that it is going to hurt do you? It just seems unlogical for a distro with such rapid updates to not have problems. I have read forums of people complaining about it and even switching to Slack because of it. But Perhaps you are right. Perhaps Ubuntu and Fedora are less stable then Arch is. I won't know because I never tried it. So agreed.
As for freebsd I have nothing against it. Its just what is going on behind the technical parts. I don't have to know or use a system to find out its politics.
Are you going to sit and tell me that FreeBSD has better Hardware support then Linux? Or that it is used by more servers then Linux. Or that some key developers did not quit the project?
All of these are facts.
I am not saying FreeBSD will come back again with better stuff. But right now this is what is going on. I actually like FreeBSD as its principles and as its system over Linux... But it is lacking support currently.
If I had the correct hardware I actually would not mind installing FreeBSD. But its my laziness that tells me that: Hey buddy, why deal with all FreeBSD problems when you have Slackware waiting for you.
I really enjoy Slack and FreeBSD doesn't seem to offer me too much more then Slack does which is why I don't have the inclination to try to install and set it up. Even if my hardware supports it.
So tell me. Am I wrong? Is FreeBSD Better then Slackware? Why? Maybe if you have some good reasons. I might try it out. I am honestly open to learn new things.
Other then Ports and ZFS what else does it offer to compete with Slackware?
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
